{"id":"W4409581635","doi":"10.31235/osf.io/wg82k_v2","title":"Updating “The Future of Coding”: Qualitative Coding with Generative Large Language Models","year":2025,"lang":"en","type":"preprint","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Canadian Institute for Advanced Research; Yale University; City University of New York; Sage Foundation","keywords":"Generative grammar; Coding (social sciences); Computer science; Natural language processing; Artificial intelligence; Linguistics; Mathematics; Statistics; Philosophy","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.06779722,0.001246991,0.0007509635,0.00302957,0.003363476,0.0103156,0.004142807,0.002496336,0.01022741],"category_scores_gemma":[0.3201284,0.001303073,0.001430798,0.002953197,0.02085018,0.0218123,0.008550578,0.006143321,0.002664869],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00712382,"about_ca_system_score_gemma":0.009165776,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005688264,"about_ca_topic_score_gemma":0.006540931,"domain_scores_codex":[0.9077734,0.07903998,0.001852862,0.004684538,0.005908056,0.0007411404],"domain_scores_gemma":[0.663354,0.2547221,0.008702617,0.05257296,0.01888944,0.001758872],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"qualitative","study_design_scores_codex":[0.0001553706,0.00005598925,0.003460994,0.000704409,0.00006524481,0.0001724834,0.0513682,0.008678955,0.001826853,0.8304624,0.01021693,0.09283217],"study_design_scores_gemma":[0.00004554481,0.00002820306,0.000518962,0.0006341161,0.00002253907,0.0001376464,0.006435598,0.03848104,0.002880752,0.9051442,0.04557294,0.00009841563],"study_design_candidate":"qualitative","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01016663,0.0003701306,0.9697312,0.009501516,0.0003090489,0.0002276725,0.0005837569,0.000924988,0.008185076],"genre_scores_gemma":[0.2963247,0.0004962559,0.6936782,0.002857901,0.0001688635,0.001288716,0.0008292748,0.001386233,0.002969937],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9322028,"threshold_uncertainty_score":0.3585503,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02330236058004553,"score_gpt":0.3402031641022996,"score_spread":0.316900803522254,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}